Dong-Hwan Lee
Hallym Institute for Clinical Medicine, Hallym University Medical Center, Anyang, Republic of Korea
Objectives: Many population pharmacokinetic (PK) models have been developed for an antibiotic, and the structures of such models are very diverse. The first aim of this study was to compare the predictability of efficacy by MCS between a true one-compartment model and a true two-compartment model for doripenem. The second aim was to explore how we can identify the usefulness of a one-compartment model when the PK/PD indices between three mis-specified one-compartments models and a true two-compartment model are compared.
Methods: The reported two-compartment model parameters of two doripenem studies and a vancomycin study were used to generate 200 virtual concentration-time profiles for each study. Sparse and dense sampling designs were selected to build the one- and two-compartment models, respectively, for the drugs. We conducted 10,000-subject simulations with the newly constructed PK models. The probability of target attainment for the PK/PD indices were compared between the one- and two-compartment models of the same drug, applying the clinical breakpoint distribution of minimum inhibitory concentration (MIC)s.
Results: The simulated concentration-time profiles well-reproduced the original data. In addition, the PTAs were similar between the one- and two-compartment models when infusion time and MIC were the same in the doripenem studies. For vancomycin simulations, the maximum difference was 65.9% between a mis-specified one-compartment model and the true two-compartment model.
Conclusions: This study explored the PK/PD indices of antibiotics using one- and two-compartment models built with the same cohort; however, with different sampling designs. When a one-compartment model was established with the sparse sampling data from a two-compartment model simulation, the probability of target attainment when evaluated by fT>MIC was similar to that of the two-compartment model. When a mis-specified one-compartment model was established using the sparse sampling data from a two-compartment model simulation, the probability of target attainment when evaluated by AUC/MIC significantly differed form that of the two-compartment model. For drugs that are commonly known to follow the two-compartment model, when a one-compartment model is developed through PK studies, the model must satisfy the following three conditions to be useful: 1. A review of the residual based diagnostic plots including the residuals vs. time plot and residuals vs. population model-prediction plot; 2. The distribution of the observation and predictions should be compared to the visual predictive check; and 3. The range of the sampling times should be wide enough based on the expected half-life, which is dependent on the drug and the condition of the patient.
Reference: PAGE 28 (2019) Abstr 8799 [www.page-meeting.org/?abstract=8799]
Poster: Drug/Disease Modelling - Infection